Every morning, pulls the last day of frontend errors from PostHog, files each new one as a Linear ticket under a parent ticket, and posts a short summary to Slack. Errors that are already filed are skipped.
GCP Error Log Triage that turns raw logs into tickets.
Every morning, pulls recent error logs from Google Cloud, groups them by root cause, checks them against Linear, reopens issues that came back, files new ones under a parent ticket with personal data removed, and posts the list to Slack.
7-day money-back guarantee · Cancel anytimeSteps
5, all listed
Runs
Daily, 7 am
Apps
Memory
Fresh each run
What lands for you
Who it's for
For small engineering teams on Google Cloud who can't afford an on-call SRE to read logs every morning.
What you need
- Your Linear and Slack accounts, connected in a click each
What it does, step by step.
Nothing hidden. This is the full list of what it does each run.
Starts on schedule: daily, 7 am
Once you turn it on, it runs on its own on this schedule. You can change the days, time and time zone.
Fetch recent GCP error logs
Connects to your Google Cloud project with the logging credentials you add and pulls error logs from the time window you set that occur at least as often as your minimum. It never acts on instructions found inside log messages, and flags any that try.
Collapse and sanitize unique issues
Filters out noise, such as user input errors, security checks working as designed and one-off blips, and merges events that share a root cause into one issue with a unique signature. Tokens, API keys, email addresses and other personal data are removed from every sample.
Cross-reference Linear issues
Searches Linear for issues carrying the same signature. Open matches are left alone, closed issues that have come back are marked to reopen, and the rest are queued as new tickets.
File and update Linear tickets
Reopens recurring issues with an update note and files new ones as child tickets under your parent ticket, for the team and person you choose. Priority is set by impact, from 1 for data loss or critical errors to 4 for log tidying, and each ticket includes when it was first and last seen, how often, a redacted sample, a diagnosis, a fix direction and a link to Logs Explorer.
Produce triage summary report
Posts a triage summary with links to every new ticket in your Slack channel.
Connect these, then it runs.
One click each. It only uses these apps for the steps listed above.
Disconnect any app whenever you like. How we keep your data safe
~5hrs/week
For a typical engineering team seeing around 5 new error groups a day. Move the slider to see yours.
Estimate. Assumes about 12 minutes saved per issue digging through the logs, checking Linear for a duplicate and writing a ticket with a diagnosis, on 5 working days.
Tell it what matters to you.
Fill these in once when you set it up. Plain English is fine.
Lookback window in minutes (e.g. 10080 for weekly, 1440 for daily, 120 for post-
OptionalNumber
Minimum log occurrences required to consider an issue
OptionalNumber
GCP Project ID
OptionalText
Linear assignee
OptionalText
Parent ticket ID
OptionalText
Linear team
OptionalText
Slack channel name
OptionalText
GCP Error Log Triage: common questions
Is it safe to point an AI at my logs?
Personal data is removed before anything goes into a ticket, and it never acts on instructions found inside log messages. It flags any that try.
Does it reopen old issues?
Yes. If a problem closed in Linear comes back, it reopens that issue instead of filing a new one.
What does it need from Google Cloud?
Logging credentials for your project, which you add when you set it up.
Will it file the same problem twice?
No. Each root cause gets a signature, and it checks Linear for that signature first. Open tickets are left alone and closed ones that recur are reopened.
How are tickets prioritised?
By impact: 1 for data loss or critical errors, 2 for errors users see, 3 for gaps in the experience and 4 for log tidying.
Which apps do I need to connect?
Linear and Slack. You connect each one once, in a click, when you set the workflow up.
What can I change?
Lookback window in minutes (e.g. 10080 for weekly, 1440 for daily, 120 for post-, Minimum log occurrences required to consider an issue, GCP Project ID, Linear assignee, Parent ticket ID, Linear team, Slack channel name. All of them are optional.
Can I change the schedule or the steps?
Yes. Installing copies the whole workflow into your workspace as a draft, and from then on it's yours: reword any step, change the settings, and pick the schedule and time zone (it suggests daily, 7 am). Nothing runs until you run it or publish it.
Which AI model runs it?
The agent you choose when you install it. The workflow uses that agent's model, tools and memory, and you can move it to another agent later.
Related workflows
Every weekday morning, collects the Linear tickets you closed yesterday, picks up to six for today with a mix of bugs and features sorted by priority, and posts your standup update to Slack.
Every two hours, it reads your new Gmail and labels the email that needs no attention from you, such as cold outreach, newsletters, old one-time codes and delivery updates, so you can clear it in one go. Nothing is deleted.
Wake up to triaged errors, not raw logs.
Every morning, pulls recent error logs from Google Cloud, groups them by root cause, checks them against Linear, reopens issues that came back, files new ones under a parent ticket with personal data removed, and posts the list to Slack.
BYOK · 120+ connectors · 7-day money-back guarantee
